Token Expand-Merge: Training-Free Token Compression for Vision-Language-Action Models

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Ye, Yifan, Ma, Jiaqi, Cen, Jun, Lu, Zhihe
Format:	Preprint
Published:	2025
Subjects:	Robotics
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

DepthCache: Depth-Guided Training-Free Visual Token Merging for Vision-Language-Action Model Inference
by: Li, Yuquan, et al.
Published: (2026)

Self-evolved Imitation Learning in Simulated World
by: Ye, Yifan, et al.
Published: (2025)

A Survey on Vision-Language-Action Models: An Action Tokenization Perspective
by: Zhong, Yifan, et al.
Published: (2025)

FAST: Efficient Action Tokenization for Vision-Language-Action Models
by: Pertsch, Karl, et al.
Published: (2025)

ProbeFlow: Training-Free Adaptive Flow Matching for Vision-Language-Action Models
by: Fang, Zhou, et al.
Published: (2026)

The Compression Gap: Why Discrete Tokenization Limits Vision-Language-Action Model Scaling
by: Shiba, Takuya
Published: (2026)

FASTer: Toward Efficient Autoregressive Vision Language Action Modeling via Neural Action Tokenization
by: Liu, Yicheng, et al.
Published: (2025)

RL Token: Bootstrapping Online RL with Vision-Language-Action Models
by: Xu, Charles, et al.
Published: (2026)

RetoVLA: Reusing Register Tokens for Spatial Reasoning in Vision-Language-Action Models
by: Koo, Jiyeon, et al.
Published: (2025)

Learning to Accelerate Vision-Language-Action Models through Adaptive Visual Token Caching
by: Wei, Yujie, et al.
Published: (2026)

IVRA: Improving Visual-Token Relations for Robot Action Policy with Training-Free Hint-Based Guidance
by: Park, Jongwoo, et al.
Published: (2026)

VQ-VLA: Improving Vision-Language-Action Models via Scaling Vector-Quantized Action Tokenizers
by: Wang, Yating, et al.
Published: (2025)

From Pixels to Tokens: A Systematic Study of Latent Action Supervision for Vision-Language-Action Models
by: Lin, Yihan, et al.
Published: (2026)

MergeVLA: Cross-Skill Model Merging Toward a Generalist Vision-Language-Action Agent
by: Fu, Yuxia, et al.
Published: (2025)

ActionCodec: What Makes for Good Action Tokenizers
by: Dong, Zibin, et al.
Published: (2026)

VLSA: Vision-Language-Action Models with Plug-and-Play Safety Constraint Layer
by: Hu, Songqiao, et al.
Published: (2025)

ECHO: Continuous Hierarchical Memory for Vision-Language-Action Models
by: Hu, Yanbin, et al.
Published: (2026)

DTP: A Simple yet Effective Distracting Token Pruning Framework for Vision-Language Action Models
by: Li, Chenyang, et al.
Published: (2026)

BFA++: Hierarchical Best-Feature-Aware Token Prune for Multi-View Vision Language Action Model
by: Li, Haosheng, et al.
Published: (2026)

ETA-VLA: Efficient Token Adaptation via Temporal Fusion and Intra-LLM Sparsification for Vision-Language-Action Models
by: Wang, Yiru, et al.
Published: (2026)

Vision-Language-Action Models for Autonomous Driving: Past, Present, and Future
by: Hu, Tianshuai, et al.
Published: (2025)

Action Tokenizer Matters in In-Context Imitation Learning
by: Vuong, An Dinh, et al.
Published: (2025)

ST4VLA: Spatially Guided Training for Vision-Language-Action Models
by: Ye, Jinhui, et al.
Published: (2026)

OAT: Ordered Action Tokenization
by: Liu, Chaoqi, et al.
Published: (2026)

VLA-Cache: Efficient Vision-Language-Action Manipulation via Adaptive Token Caching
by: Xu, Siyu, et al.
Published: (2025)

RynnVLA-002: A Unified Vision-Language-Action and World Model
by: Cen, Jun, et al.
Published: (2025)

TIC-VLA: A Think-in-Control Vision-Language-Action Model for Robot Navigation in Dynamic Environments
by: Huang, Zhiyu, et al.
Published: (2026)

BLURR: A Boosted Low-Resource Inference for Vision-Language-Action Models
by: Ma, Xiaoyu, et al.
Published: (2025)

GST-VLA: Structured Gaussian Spatial Tokens for 3D Depth-Aware Vision-Language-Action Models
by: Sarowar, Md Selim, et al.
Published: (2026)

TTF-VLA: Temporal Token Fusion via Pixel-Attention Integration for Vision-Language-Action Models
by: Liu, Chenghao, et al.
Published: (2025)

Focusing on What Matters: Object-Agent-centric Tokenization for Vision Language Action models
by: Bendikas, Rokas, et al.
Published: (2025)

Grid-based Fast and Structural Visual Odometry
by: Zhihe, Zhang
Published: (2024)

Robust Finetuning of Vision-Language-Action Robot Policies via Parameter Merging
by: Yadav, Yajat, et al.
Published: (2025)

VP-VLA: Visual Prompting as an Interface for Vision-Language-Action Models
by: Wang, Zixuan, et al.
Published: (2026)

History-Conditioned Spatio-Temporal Visual Token Pruning for Efficient Vision-Language Navigation
by: Wang, Qitong, et al.
Published: (2026)

RLRC: Reinforcement Learning-based Recovery for Compressed Vision-Language-Action Models
by: Chen, Yuxuan, et al.
Published: (2025)

Co-Me: Confidence-Guided Token Merging for Visual Geometric Transformers
by: Chen, Yutian, et al.
Published: (2025)

RoVLA: Multi-Consistency Constraints for Robust Vision-Language-Action Models
by: Luo, Jingzhou, et al.
Published: (2026)

A Hierarchical Spatiotemporal Action Tokenizer for In-Context Imitation Learning in Robotics
by: Fateh, Fawad Javed, et al.
Published: (2026)

GraphCoT-VLA: A 3D Spatial-Aware Reasoning Vision-Language-Action Model for Robotic Manipulation with Ambiguous Instructions
by: Huang, Helong, et al.
Published: (2025)